作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (17): 202-204. doi: 10.3969/j.issn.1000-3428.2008.17.072

• 人工智能及识别技术 • 上一篇    下一篇

常用边缘检测算法的定量比较

陈彦燕,王元庆   

  1. (南京大学电子科学与工程系,南京 210093)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-09-05 发布日期:2008-09-05

Quantitative Comparison of Common Edge Detection Algorithms

CHEN Yan-yan, WANG Yuan-qing   

  1. (Depterment of Electronic Science & Engineering, Nanjing University, Nanjing 210093)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-09-05 Published:2008-09-05

摘要: 在总体定性分析的基础上,根据边缘连通度、抗噪性能、运算复杂度等典型的边缘检测算法性能标准,利用定量计算结果比较并验证各种边缘特征提取算法的优缺点。采用B样条三次曲线对原始图像进行拟合,显著提高了检测后的图像质量。实验结果表明,该方法可以很好地应用于图像处理领域。

关键词: 边缘检测, 边缘连通度, 频谱图, 三次样条插值

Abstract: Based on whole qualitative analysis, this paper compares and validates merits and demerits of various edge detection algorithms in a quantificational way from several typical judgement standards for edge detection algorithms like edge connectivity, nosie resistence ability, operation complexity. By means of B cubic spline interpolation, it improves the quality of the detected images. Experiental results show that this methods can be well applied in the region of image processing.

Key words: edge detection, edge connectivity, spectrum map, cubic spline interpolation

中图分类号: